Many studies faced the problem of robot autonomous navigation in different fields, but nowadays just a few of them uses all the implicit information coming from the context in which such navigation is occurring. This results in a huge potential information loss that prevents us from adapting the robot's behaviour to each different situation it may be in. In this paper we therefore define the Context of Navigation using a semantic model built with ontologies. Then, we show what kind of inference is useful in such application. Finally, we provide a method to exploit the information coming from the context together with the control loop of the robot to better navigate it. Two application examples are provided to illustrate these concepts: a service robot and an intelligent autonomous vehicle.